Light tracking Solar pannel

[lompa]

for my project i was thinking of making a solar pannel control system that will track the light and move the panel towards the strongest source of light. I am looking for advice on how i would implement this, i was thinking mounting the panel onto a mechanical device that would move the panel via x and y axis.

I was thinging using sevos to move the panel and use LDR's to sense for the light intensitity, and use a PLC as a control system.

Any advice on were i am going wrong or any ideas how this would be improved would be very much appreciated, thanks for your time

 

[-=GST=- Nemisis (cs/cz)]

have a LDR (light dependent resistor i think) then have a chip or sub circuit that detects the strength of the current and the if it lower then a certain level to move motor on axis till the currents is at its highest. this isnt exactly precision but its cheapish and would work ok.

 

[stevez]

I'd be inclined to do the tracking based on time. On a sunny day the control system requirements would be reasonably simple. Suppose you had a day when the sun came up in the morning, the solar panel was pointed that way and then it got cloudy until afternoon. Would your tracking system have a large enough field of view to see the sun when it came out on the other side of the sky? This large enough field of view might also apply to an optical tracking system - would it be able to see the sun when it came up in the morning, given that it stopped seeing the sun in the opposite direction the day before?

Another thing I wonder about - if the tracking system had a broad field of view on a partly cloudy day, would it start tracking other 'bright' spots in the sky when the sun was behind a cloud. When the sun came back out it should go back but would you be running the motors an awful lot. With a narrow field of view the problem could be worse.

It might be that a combination of time plus some optical fine tuning would yeild the optimum. Another thought is that a single axis movement might yeild lots of benefits - depending on where you live. The sun tracks across the sky within two limits thru the year. Maybe time controls one axis and using optical to address the elevation changes thru the year. That would limit the possible places for the sun to be at any given time.

 

[grrr_arrghh]

not so long ago, I was shown a set-up for projecting an image of the sun onto a screen (for the purposes of seeing sun-spots). This (obviously) used a mirror, but a feature of the mirror was that it had a motor that rotated the mirror at a certain angle so that it tracked the sun across the sky. I suspect the mirror was also parabolic.

I only mention this because looking at astronomy websites/shops may yield some useful suggestions/info.

Tim

 

[Klaus]

for my project i was thinking of making a solar pannel control system that will track the light and move the panel towards the strongest source of light. I am looking for advice on how i would implement this, i was thinking mounting the panel onto a mechanical device that would move the panel via x and y axis.

I was thinging using sevos to move the panel and use LDR's to sense for the light intensitity, and use a PLC as a control system.

Any advice on were i am going wrong or any ideas how this would be improved would be very much appreciated, thanks for your time

If this is going to be a practical project, make sure you do your sums on net power gained by tracking and power consumed by the tracking system.
You might find that for a small panel you gain very little or even lose power.

The whole idea of this is to make a panel more efficient :wink:

The other pitfall is the actual mechanics of the tracking. Be sure you sort this out to work reliably and with as little friction as possible in all the weather conditions foreseen at your site. No good to spend $ on fancy tracking machinery when the first puff of wind renders it useless
Klaus

 

[lompa]

thanks for all your help and time, could anyone recommend the best kind of light sensor to use for this project, and would it be more practical to use a PIC to control the motors?


also if say 1 sensor picks up the strongest source of light and the pannel moves in that direction, how would i make the pannel stop moving at the optimum position?

 

[mechie]

Tracking the Sun

Would something like this work :?: (attached drawing)

If a small rectangular box was constructed as below...
a central compartment aligned with a slot in the lid will catch the light when aligned with the sun - this is lined with black felt (or painted matt black) to prevent any light escaping.
To either side of this is a compartment containing some form of sensor, LDR, photovoltaic, whatever). This has silvered sides to maximise light reacing the sensor regardless of its alignment. The inside of the roof is matt black to minimise light spill to the other compartments.

As the sun crosses the sky so the light reaching one cell will exceed that reaching the other. A difference signal will cause a corrective drive to realign the detector (and solar panel) with the sun.
If the sun disappears (thunder storm or overnight) the detector will stay at the last known good position (unless a bright star or street light attracts it). When the sun reappears there could be a major alignment error but the silvered sides of the box allows for a very wide acceptance angle so the detector will still see the error and the position controller will start to correct the alignment.

A complete second detector could be used for the second axis to allow for elevation tracking.

 

[stevez]

Mechie's sensor looks interesting. If on the first cycle, it followed the sun properly (seems like it will) you'd need a way to get it to be able to see the sunrise unless you added some time of day intelligence. It would seem that a PIC could provide that.

As already suggested, the gain from tracking the sun precisely as compared to no tracking or roughly tracking may not be great. The photocells, solar cells or whatever you might use in Mechie's sensor and the associated electronics are relatively inexpensive if you shop carefully. Maybe a crude, fixed sensor array (multiples of Mechie's) and some logic in a PIC can be used to crudely track the sun or the brightest spot in the sky on cloudy days.

 

[lompa]

i will most likely, start with an west-east tracking system first, then if all goes well i will encorporate tracking for north-south (for when the seasons change).

I have been looking at some panels on ebay, the project is to demonstrate the idea but if possible i would like to power the whole system of just the solar panel(depending on cost). I will most likely use a RC servo motor and a PIC Microcontroller system.

will this panel provide enough power or would i need a really big one.

**broken link removed**

 

[lompa]

Also i have discovered that when the sun goes down the pannel will be stuck in the wrong position, therefore i would need a sensor and a battery to reset the panel for the morning.

Im planning on running the PIC and Servo off the pannel and at night to reset the pannel 4 AA rechargaeble batteries. these would need to be charged by the pannel.

does anyone have any info on solar rechargers that are suitable, for a AA batteries or a similar voltage battery?

 

[williB]

i have one of those , mine is an ICP global technologies..
with no load it produces 18.48 V..

 

[williB]

although it says it will produce 324 mA..i have never gotten that much from it..
with my Amp meter across its output it only produces 110 mA in direct sunlight.
the meter has an internal resistance of 6.3 ohms on the amp scale.

 

[Nigel Goodwin]

although it says it will produce 324 mA..i have never gotten that much from it..
with my Amp meter across its output it only produces 110 mA in direct sunlight.
the meter has an internal resistance of 6.3 ohms on the amp scale.

As I've often said before, the specs on solar panels are the absolute maximum - think sahara desert at noon! 8)

 

[jbeng]

Question: does the output of a solar panel vary greatly if the sunlight strikes the panel at a 45 degree angle vs a 90 degree angle (perpendicular)? If not, then the panel will not have to move nearly as much, reducing the power consumed by the tracking system. Also, it will possibly eliminate the need for the mechanical system to drive it in the Y-axis altogether. For example: if the sun is directly overhead at noon on the summer solstice and at a 45 degree angle in the sky at noon on the winter solstice, then why not set the Y-axis angle at 67.5 degrees? With the panel at that angle, the sunlight would be at maximum only +/-22.5 degrees from hitting the panel in a perpendicular fashion. Does this make any sense? Keep in mind that I have very little experience with photovoltaic panels.
JB

 

[williB]

i am not sure by how much the output will vary , but it will not be optimum.
what would be so hard to adjust the notth south angle every week or so..
what is your latitude lompa ?

 

[Nigel Goodwin]

Question: does the output of a solar panel vary greatly if the sunlight strikes the panel at a 45 degree angle vs a 90 degree angle (perpendicular)?

I suggest you try it, position the panel at the optimum angle and measure the output, then try angling it either way and see how much it drops by.

From a purely practical point of view, you see LOAD's of solar panels powering devices at the side of the road - none of these are steerable, which makes you think it's not a practical solution!. Certainly you need to make a detailed study of how much a fixed panel will provide, and how much a steered one will (you could steer it by hand every hour or so). From that you can calculate how much extra it would provide, then decide if the energy cost of steering it would have a positive or negative effect on performance.

I visited C.A.T. (Centre for Alternative Energy!) in Wales a few years back, they had banks of solar panels - none of them steered!.

 

[williB]

ok priliminary results of the solar panel question..

conditions : Brite sunny morning no clouds at all
time: 9:30
solar panel: icp global tech : battery saver plus - different from previously mentioned panel..
Internal resistencce of meter : 1.8 ohms .

panel sittong on window sill : 15.1 mA
panel at approx 45 degrees : 36.0 mA
panel pointing directly at sun :39.0 mA

 

[Nigel Goodwin]

panel at approx 45 degrees : 36.0 mA
panel pointing directly at sun :39.0 mA

Hardly seems worth bothering with steering for 3mA?, only an 8% increase.

 

[williB]

yeah but look at the 61.5 percent increase from the pannel just sitting on the sill. pointing straight up.
as opposed to pointing tward the sun..

 

[williB]

Also i have discovered that when the sun goes down the pannel will be stuck in the wrong position, therefore i would need a sensor and a battery to reset the panel for the morning.

Im planning on running the PIC and Servo off the pannel and at night to reset the pannel 4 AA rechargaeble batteries. these would need to be charged by the pannel.

does anyone have any info on solar rechargers that are suitable, for a AA batteries or a similar voltage battery?

i think the whole idea of solar power is to charge a battery that u can use while it is sunny out , or later on when it is dark.